1

Simulation of Boundary Layer

Ingestion Fan Noise for NOVA Aircraft

Configuration

G.Romani, Q.Ye, F.Avallone, D.Ragni, D.Casalino

TU-Delft/3DS Workshop on PowerFLOW Simulations of Aircraft Noise

14th September 2018, Delft

2

Outline

• Background and motivation

• Numerical method

• Geometries and simulated cases

• Computational setup

• Numerical results

• Conclusions and future outlooks

3

Outline

• Background and motivation

• Numerical method

• Geometries and simulated cases

• Computational setup

• Numerical results

• Conclusions and future outlooks

4

Background and motivation

• UHBR engines on next generation aircraft Address increasingly stringent

aviation regulations for pollution and noise impact

Enhanced propulsion efficiency and lower noise emissions

Integration challenges and special designs required

• Four different NOVA (Nextgen Onera Versatile Aircraft) aircraft geometries

investigated at Onera with focus on engine integration options*

*Wiart et al., “Development of NOVA Aircraft Configurations for Large Engine Integration Studies“, AIAA 2015-2254

5

Background and motivation

• Boundary Layer Ingestion (BLI) configuration benefits

Mass and drag penalty reduction

Jet and wake losses reduction

• Many implications have to be addressed before deriving the associated

benefits: effects of inlet flow distortion on engine efficiency, operability,

aeromechanics and aeroacoustics

• Research goals:

To perform the first CFD/CAA simulation of a full aircraft+BLI fan stage system

To address BLI installation effects on fan noise for a NOVA BLI-like configuration

Potential fuel burn reduction

6

Outline

• Background and motivation

• Numerical method

• Geometries and simulated cases

• Computational setup

• Numerical results

• Conclusions and future outlooks

7

Numerical method

• SIMULIA PowerFLOW solver*:

– Lattice-Boltzmann method for subsonic/supersonic flows

• Solves the fully explicit, transient and compressible LBE

• Sliding mesh (LRF) for rotating geometries

• Hybrid solver: D3Q39 model inside LRF, D3Q19 model outside LRF

– LBM-VLES turbulence model

• Large-eddies are resolved (“coherent” statistically anisotropic eddies)

• Small eddies (statistically universal) are modeled with an extended RNG k-ε model

• Swirl term used to switch from modeled to resolved eddies

– Extended turbulent wall model to account for favorable/adverse pressure gradients

Hybrid D3Q39 ModelMamax~2.0

*Nie et al., “A Lattice-Boltzmann/Finite-Difference Hybrid Simulation of Transonic Flow“, AIAA 2009-139Gonzalez-Martino et al., “Fan Tonal and Broadband Noise Simulations at Transonic Operating Conditions Using Lattice-Boltzmann Methods“, AIAA 2018-3919van der Velden et al., “Jet Noise Prediction: Validation and Physical Insight“, AIAA 2018-3617

8

Outline

• Background and motivation

• Numerical method

• Geometries and simulated cases

• Computational setup

• Numerical results

• Conclusions and future outlooks

9

Fan stage geometry

• Modified version of “Low-Noise” NASA/SDT*– Original geometry fully scaled to match NOVA fan diameter (2.15 m)

– Original nacelle axial length increased to match NOVA BLI intake-fan distance (2.35 m)

*Envia , “Fan Noise Source Diagnostic Test - Vane Unsteady Pressure Results”, AIAA 2002-2430

Fan (22 blades)

OGV (26 stator swept vanes)

ø = 2.15 m

Tip clearance = 1.94 mm

10

• Engine integration on NOVA lifting fuselage (courtesy of ONERA)– BLI engine

– 40% buried intake

– Intake-Fan distance = 2.35 m

– Tilt angle = 1°

– Toe angle = 2.5°

Fan stage integration into NOVA fuselage

ONERA s-duct

design costraints

S-duct design

Semi-span = 19.05 mFuselage length = 38.3 m

11

Simulated cases

• Isolated NASA/SDT with modified nacelle

• Installed NASA/SDT with modified nacelle into NOVA fuselage geometry

Mach Mach Tip Pressure Temperature AoA Glide angle Tilt angle Toe angle

0.25 1.0038 ISA at 1000 ft 4° 6° 1° 2.5°

12

Outline

• Background and motivation

• Numerical method

• Geometries and simulated cases

• Computational setup

• Numerical results

• Conclusions and future outlooks

13

Permeable Surface for FWHVR strategy

Computational setup

*Gonzalez-Martino et al., “Fan Tonal and Broadband Noise Simulations at Transonic Operating Conditions Using Lattice-Boltzmann Methods“, AIAA 2018-3919

Grid

Resolution

Fan Tip Cell

Size (mm)# Cells CPUh (10 revs)

Medium* 0.355 611 M 56000

Local Reference Frame

LRF

• Symmetry plane (at fuselage centerline)

• FWH permeable approach for far-field noise

• 16 Variable Resolution (VR) regions (medium

grid resolution)– VR16: tip gap

– VR15: leading/trailing edges of fan/OGV and nacelle lip

– VR14: fan/OGV

– VR13: bypass channel, nacelle and s-duct walls

– VR12: FWH permeable surface

– VR11-VR0: fuselage offsets and boxes up to domain

boundaries

• Fan geometry rotated through LRF

14

Outline

• Background and motivation

• Numerical method

• Geometries and simulated cases

• Computational setup

• Numerical results

• Conclusions and future outlooks

15

• Instatanoeus flow on a plane normal to the fuselage– Higher flow acceleration at intake lip

– Adverse pressure gradient induced flow separation on intake wall

– Adverse pressure gradient induced flow separation on s-duct surface

– Different fan wake/OGV interaction

Flow installation effects

Isolated engine BLI engine

17

• Mean flow on a plane upstream the fan– Strong flow distortion and non-uniformity

– Flow acceleration in blade inboard regions

Flow installation effects

Isolated engine BLI engine

0°

90°

180°

270°

0°

90°

180°

270°

Fan blade azimuth

18

Fan blade sectional air-loads

• Phase-locked cxŨ2 at three fan blade span-wise locations

– Inboard: low-frequency unsteadiness mean flow distortion

– Outboard: high-frequency unsteadiness and lower mean value turbulence ingestion

Blade Inboard Blade Outboard

Ũ =u

a∞

Local flow velocity

Freestream speed of sound

Negative Cx as the

meaning of thrust

22

Arc 0°

• PSD on 10 m radius arc centered around the fan center

Far-field noise directivity - Arc 0°

FLOW

FLOW

FLOW

BLI

en

gin

eIs

ola

ted

en

gin

e

23

• PSD on 10 m radius arc centered around the fan center Arc 45°

Far-field noise directivity - Arc 45°

FLOW

FLOW

FLOW

BLI

en

gin

eIs

ola

ted

en

gin

e

24

• PSD on 10 m radius arc centered around the fan center

Arc 90°

Far-field noise directivity - Arc 90°

FLOW

FLOW

FLOW

BLI

en

gin

eIs

ola

ted

en

gin

e

25

Arc 45°

Far-field noise – Mic at 20°/Arc 45°

FLOW

• PSD for directivity angle of 20° on Arc at 45°

26

Arc 45°

Far-field noise – Mic at 90°/Arc 45°

FLOW

• PSD for directivity angle of 90° on Arc at 45°

27

Arc 45°

Far-field noise – Mic at 160°/Arc 45°

FLOW

• PSD for directivity angle of 160° on Arc at 45°

28

• CFD computed pressure waves around BPF1– Isolated engine: low/high pressure areas extending upstream from each fan blade and co-rotating

with fan propagate mainly upstream in the sideline direction

– BLI engine: highly irregular pressure waves pattern propagating mainly upstream in the axial

direction and downstream in the sideline direction

Band-pass filtered pressure around BPF1

Isolated engine BLI engine

30

PNL on-the-ground

Isolated engine BLI engine

• Perceived Noise Level vs time during a takeoff fight path

33

Outline

• Background and motivation

• Numerical method

• Geometries and simulated cases

• Computational setup

• Numerical results

• Conclusions and future outlooks

34

Conclusions and future outlooks

• LBM solver Simulia PowerFLOW used to address fan noise implications of

NOVA BLI-like engine configuration

• Aerodynamic installation effects:– Flow distortion and non-uniformity low-frequency air-loads variation

– Separation on intake and s-duct walls high-frequency air-loads variation

• Aeroacoustic installation effects:– Increase of noise sources intensity, but different propagation behavior

• Isolated engine: noise radiated mainly upstream in the sideline direction

• BLI engine: noise radiated mainly upstream in the axial direction and downstream in the sideline direction

• As future outlooks:– Analysis of boundary layer/fan interaction mechanisms

– Analysis of fan wake/OGV interaction mechanisms

35

Thank you for your attention!

This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant No 769 350